Establishment of a prediction model for the deterioration of renal function during hospitalization in the patients with acute heart failure

Cheng Lingling; Yao Jiaping; Ji Xixi

doi:10.3969/j.issn.1002-1949.2025.04.007

Abstract：Objective To analyze the related factors that may affect the deterioration of renal function in the patients with acute heart failure during hospitalization and to establish a prediction model based on this,so as to provide references for early clinical identification of renal function deterioration.Methods The clinical data of 398 patients with acute heart failure admitted to the emergency department of Zhejiang Provincial People's Hospital from January 2021 to December 2023 were retrospectively analyzed.Based on the occurrence of renal function deterioration during hospitalization,the patients were divided into deterioration group(n=79)andnon-deteriorationgroup(n=319).The general data,relevant treatment measures and laboratory indicators during hospitalization were compared between the two groups.Lasso regression and Logistic regression analysis were used to screen the related factors that may affect the deterioration of renal function in the patients with acute heart failure.Based on these factors,a prediction model was constructed,and the prediction efficiency of the model was evaluated by decision curve,nomogram and calibration curve.Results Compared with the non-deterioration group,the patients in the deterioration group had the higher proportion of hypertension,no use of sodium-dependent glucose transporters 2(SGLT-2)inhibitors,daily doses of furosemide≥80 mg,the higher levels of N-terminal pro-brain natriuretic peptide(NT-proBNP)and creatinine on admission,and the lower hemoglobin levels upon admission(P<0.05).Lasso regression and Logistic regression analysis showed that the above-mentioned factors may all be risk factors for the deterioration of renal function in the patients with acute heart failure(P<0.05).Based on the results of regression analysis,the prediction model was constructed,and the decision curve of the model was drawn.It was found that when the threshold was in the range of 0.00-0.93,the clinical net benefit rate of the model was always greater than 0,and the maximum net benefit rate was 0.183.The nomogram was drawn to visualize the risk of renal function deterioration in the patients.The results showed that the C-index of the model for predicting renal function deterioration in the patients with heart failure was 0.945(95%CI 0.914-0.973),and the calibration curve approached the ideal curve.Conclusions The predictive model constructed based on the hypertension,SGLT-2 inhibitor use,daily dose of furosemide,hemoglobin,NT-proBNP and creatinine levels on admission has ideal predictive performance for the risk of renal function deterioration in acute heart failure patients.